Skip to content

Conversation

@shino16
Copy link
Collaborator

@shino16 shino16 commented Nov 20, 2025

Part of fix for #2677. As explained in #2677 (comment), this patch fixes an error which is caused by feeding a NumberProxy for shape directly into fd.ops.full.

fd: FusionDefinition,
lc_to_nv_map: dict,
) -> Any:
nv_shape = getnv(shape, fd, lc_to_nv_map, inline_number=True)
Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

In my interpretation, we put inline_number=True simply because we don't want to create fd.define_scalar for each dimension every time.

Copy link
Collaborator

@beverlylytle beverlylytle left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yay!

Copy link
Collaborator

@crcrpar crcrpar left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can we have a test for this?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants